bioRxiv preprint doi: https://doi.org/10.1101/320317; this version posted May 11, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 1 Full Title: A community-level investigation of the yellow fever virus outbreak in South Omo 2 Zone, South-West Ethiopia, 2012-2014 3 4 Short Title: Investigation of a recent yellow fever outbreak in Ethiopia 5 6 Authors: 7 Ranya Mulchandani1, Fekadu Massebo2, Fekadu Bocho2, Claire L Jeffries1, Thomas Walker1, 8 Louisa A Messenger1* 9 10 Author Affiliations: 11 1Department of Disease Control, Faculty of Infectious Tropical Diseases, London School of 12 Hygiene and Tropical Medicine, UK 13 2Department of Biology, Arba Minch University, Ethiopia 14 15 *Corresponding author 16 17 Abstract (250-300 words) 18 Background 19 A yellow fever (YF) outbreak occurred in South Omo Zone, Ethiopia in 2012-2014. This study 20 aimed to analyse historical epidemiological data, to assess the risk for future YF outbreaks 21 through entomological surveillance, including mosquito species identification and molecular 22 screening for arboviruses, and finally to determine the knowledge, attitudes and current 23 preventative practices within the affected communities. 24 25 Methodology/Principal Findings 26 From October 2012 to March 2014, 165 cases and 62 deaths were reported, principally in rural 27 areas of South Ari region (83.6%), south-west Ethiopia. The majority of patients were 15-44 28 years old (74.5%) and most case deaths were males (76%). Between June and August 2017, 29 688 containers were sampled from across 177 households to identify key breeding sites for 30 Aedes mosquitoes. Ensete ventricosum (“false banana”) was identified as the primary natural 31 breeding site, and clay pots outside the home as the most productive artificial breeding site. 32 Entomological risk indices from the majority of sites were classified as “high risk” for future 33 outbreaks under current World Health Organization criteria. Adult trapping resulted in the 34 identification of members of the Aedes simpsoni complex in and around households. Screening 35 of adult females revealed no detection of yellow fever virus (YFV) or other arboviruses. 88% 36 of 177 participants had heard of YF, however many participants easily confused transmission 37 and symptoms of YF with malaria, which is also endemic in the area. 38 39 Conclusions/Significance 40 Study results emphasise the need for further entomological studies to improve our 41 understanding of local vector species and transmission dynamics. Disease surveillance systems 42 and in-country laboratory capacity also need to be strengthened to facilitate more rapid 43 responses to future YF outbreaks. 44 45 Author Summary (150-200 words) 46 Despite the availability of a highly effective vaccine, yellow fever virus (YFV) remains an 47 important public health problem across Africa and South America due to its high case-fatality 48 rate. This study aimed to assess and reduce the risk for future outbreaks. During this study, 49 historical data analysis was conducted to understand the epidemiology of the recent outbreak 50 in 2012-2014. Entomological surveillance was also carried out, including both mosquito bioRxiv preprint doi: https://doi.org/10.1101/320317; this version posted May 11, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 51 species identification and molecular screening for arboviruses, as well as a household survey 52 to understand the knowledge and attitudes towards yellow fever (YF) within the affected areas 53 and to assess community-level practices for YF prevention. We found a high abundance of 54 Aedes simpsoni complex in the context of low vaccination coverage. Community knowledge 55 and practice levels were low for reducing potential breeding sites, highlighting the need for 56 increased dissemination of information to community members on how to reduce their risk of 57 exposure to mosquito vectors of arboviruses. 58 59 Introduction 60 Yellow fever virus (YFV) is a flavivirus transmitted primarily to humans and non-human 61 primates through the bite of an infected female Aedes spp. or Haemagogus spp. mosquito (1). 62 YFV is endemic to Africa and Latin America where it causes a spectrum of clinical symptoms 63 ranging in severity from asymptomatic infection, mild illness with flu-like symptoms to severe 64 disease including, fever, jaundice or haemorrhage and death (2). Despite the availability of a 65 highly effective vaccine, YFV continues to occur in epidemic situations, and it is estimated to 66 result in 130,000 human cases and 78,000 deaths annually in Africa alone (2). Since the 1980s, 67 there has been an unprecedented rise in the number of large YFV outbreaks (3), including 68 Angola and the Democratic Republic of Congo in 2016 which together became one of the 69 largest outbreaks in Africa for more than 20 years (4–6). In early 2018, numerous cases were 70 confirmed in Nigeria and Brazil, which was particularly concerning as they were being reported 71 from areas previously not considered at-risk (7). YFV has also re-emerged across East Africa 72 with outbreaks in Ethiopia in 2012 and Uganda in 2016 (8). Due to this global resurgence, YFV 73 continues to be considered a significant threat to public health (9). 74 The resurgence of epidemic yellow fever (YF) and the increased risk of urban outbreaks is 75 multi-factorial (10,11). Rapid urbanisation, population migration, climatic changes and 76 increased travel have all been implicated in expanding the geographical range of YFV and 77 driving mosquito vectors closer to human dwellings where unvaccinated individuals are often 78 living in highly populated areas (12). A multi-faceted approach is necessary for YFV control, 79 which includes strong laboratory and surveillance systems with rapid case reporting, 80 appropriate case management, vector control and reactive and preventive vaccination 81 campaigns (13). 82 Ethiopia has experienced numerous YFV outbreaks since the 1940s. Between 1960-1962, the 83 largest YFV outbreak ever recorded in Africa occurred along the River Omo, Ethiopia (Gamo 84 Gofa, Jinka and Kaffa regions) which resulted in approximately 200,000 human cases and 85 30,000 deaths. In 1966, YFV appeared in Arba Minch, in an area previously unaffected in the 86 1960 epidemic and therefore excluded from the mass vaccination campaign at the time. During 87 this outbreak, 450 deaths were reported (2,200 human cases) and the outbreak was confirmed 88 through serological and entomological testing (14). After almost a 50-year absence, YFV re- 89 emerged in 2012. A reactive vaccination campaign commenced in June, 2013, which reached 90 approximately 550,000 people across the at-risk population. There have been no further 91 vaccination campaigns since this outbreak, and Ethiopia is one of the few remaining endemic 92 countries that have not introduced the YF vaccine into their childhood immunization 93 programme. Therefore, the country is classified as a top priority through the Eliminating 94 Yellow Fever Epidemics (EYE) strategy, a coalition of partners led by the World Health 95 Organization (WHO), UNICEF and Gavi, the Vaccine Alliance (15). 96 Knowledge, Attitudes and Practices (KAP) studies have been widely used to understand the 97 community context of disease transmission, to help inform appropriate control and risk bioRxiv preprint doi: https://doi.org/10.1101/320317; this version posted May 11, 2018. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY 4.0 International license. 98 communication activities, with an overall aim to reduce barriers to the prevention of infectious 99 diseases. In Ethiopia, there is a considerable paucity of such data on YFV, nor had a YFV 100 outbreak occurred in the region for almost 50 years. To guide appropriate, prospective disease 101 control interventions, this study aimed to collect historical epidemiological information of the 102 2012 YFV outbreak, to assess the risk for future outbreaks through entomological surveillance, 103 including mosquito species identification and molecular screening for arboviruses, and finally 104 to determine knowledge and attitudes towards YF within the affected communities following 105 this outbreak and assess current community-level practices for YF prevention. 106 Methods 107 108 Study location 109 The study was conducted in South Omo Zone (SOZ), Ethiopia, which is located in south-west 110 Ethiopia (Southern Nations Nationalities and People’s Region – SNNPR) across 5 selected 111 kebeles (villages), between June and August 2017 (Figure 1). Aykamer, Shepe, Arkisha (South 112 Ari woreda (region)), Hana (Salamago woreda) and Besheda (Hammer woreda) kebeles were 113 selected as representative sites which reported varying numbers of cases during the 2012-2014 114 outbreak and were all targeted for vaccination during the emergency reactive campaign in 115 2013. 116 Figure 1: Map of study sites in South Omo Zone (SOZ), Southern Nations Nationalities and 117 People’s Region (SNNPR), south-west Ethiopia. 118 Historical clinical data collection 119 In November 2012, cases of an unknown febrile illness were reported to the Public Health 120 Emergency Management (PHEM) through the mandatory weekly reporting format for health 121 extension workers (HEW) on all immediately reportable diseases. Symptoms were a 122 combination of fever, headache, nausea, bloody vomiting, abdominal pain, joint pain and 123 jaundice. A team from the SOZ Health Department, WHO Ethiopia Country Office and 124 Ethiopian Public Health Institute (EPHI) were deployed to the field for a rapid risk assessment.